Situation
The company: At MTS, the leading telecom company with 88 million users and a large digital ecosystem, I was a part of the MyMTS team.
The product portfolio:
MyMTS (30+ million MAU) is a self-service app. It was the #1 mobile app and #30 among all Russian apps (by AppStore ranking in 2018). MyMTS app is accompanied by the self-service websites: mts.ru and lk.mts.ru (10+ million MAU).
I managed a portfolio of these 3 B2C web and mobile products within the “Loyalty, ads, and targeted communications” vertical.
Context:
My boss (who was 2 levels above me) told me to implement one feature. He believed that it would simplify the product.
Task
I have been asked to implement a feature to simplify the integration of a loyalty program to the self-serve platforms. The key metrics that the feature was targeting were:
- The sign-ups for the loyalty program from the platforms,
- The increase in conversion rate (the number of purchases through the platforms),
- The LTV boost for the activated users.
Action
As far as I have been deeply immersed in customer discovery, I had to ask the boss 3 times to allow me to verify his hypothesis instead of starting to implement it right away. From what I knew from previous user research and data analysis, the integration and the loyalty program itself were already hard to understand and use. Considering that the idea of my boss implied massive changes, I wanted to make sure that we understand what we are doing.
I analysed that the suggested feature would be extremely technically complex and would require extensive changes in the billing systems, the loyalty program, the self-service platforms, and also in the communication channels (chat, SMS).
I saw a risk that the change could bring more costs than profit, which could mean that its impact on LTV would be rather negative than positive. We could have had some immediate increase in sign-ups and CVR (conversion rate), but the resulting long-term profit would be more significant than those short-term wins.
Results
User research revealed that the loyalty program had been difficult to understand already, so together with counting all the possible software development expenses it became clear that the hypothesis should have been rejected.
I’m proud that I didn’t just implement the feature without proper research and that I stood up for the users and for the product’s profitability.
Period
Around 2018-2019
